Innovation Labs in Corporations

𝗕𝘂𝗶𝗹𝗱𝗶𝗻𝗴 𝗮 𝗦𝗰𝗮𝗹𝗮𝗯𝗹𝗲 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁 𝗜𝘀𝗻’𝘁 𝗝𝘂𝘀𝘁 𝗔𝗯𝗼𝘂𝘁 𝘁𝗵𝗲 𝗠𝗼𝗱𝗲𝗹 — 𝗜𝘁’𝘀 𝗔𝗯𝗼𝘂𝘁 𝘁𝗵𝗲 𝗔𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲. In the age of Agentic AI, designing a scalable agent requires more than just fine-tuning an LLM. You need a solid foundation built on three key pillars: 𝟭. 𝗖𝗵𝗼𝗼𝘀𝗲 𝘁𝗵𝗲 𝗥𝗶𝗴𝗵𝘁 𝗙𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸 → Use modular frameworks like 𝗔𝗴𝗲𝗻𝘁 𝗦𝗗𝗞, 𝗟𝗮𝗻𝗴𝗚𝗿𝗮𝗽𝗵, 𝗖𝗿𝗲𝘄𝗔𝗜, and 𝗔𝘂𝘁𝗼𝗴𝗲𝗻 to structure autonomous behavior, multi-agent collaboration, and function orchestration. These tools let you move beyond prompt chaining and toward truly intelligent systems. 𝟮. 𝗖𝗵𝗼𝗼𝘀𝗲 𝘁𝗵𝗲 𝗥𝗶𝗴𝗵𝘁 𝗠𝗲𝗺𝗼𝗿𝘆 → 𝗦𝗵𝗼𝗿𝘁-𝘁𝗲𝗿𝗺 𝗺𝗲𝗺𝗼𝗿𝘆 allows agents to stay aware of the current context — essential for task completion. → 𝗟𝗼𝗻𝗴-𝘁𝗲𝗿𝗺 𝗺𝗲𝗺𝗼𝗿𝘆 provides access to historical and factual knowledge — crucial for reasoning, planning, and personalization. Tools like 𝗭𝗲𝗽, 𝗠𝗲𝗺𝗚𝗣𝗧, and 𝗟𝗲𝘁𝘁𝗮 support memory injection and context retrieval across sessions. 𝟯. 𝗖𝗵𝗼𝗼𝘀𝗲 𝘁𝗵𝗲 𝗥𝗶𝗴𝗵𝘁 𝗞𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲 𝗕𝗮𝘀𝗲 → 𝗩𝗲𝗰𝘁𝗼𝗿 𝗗𝗕𝘀 enable fast semantic search. → 𝗚𝗿𝗮𝗽𝗵 𝗗𝗕𝘀 and 𝗞𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲 𝗚𝗿𝗮𝗽𝗵𝘀 support structured reasoning over entities and relationships. → Providers like 𝗪𝗲𝗮𝘃𝗶𝗮𝘁𝗲, 𝗣𝗶𝗻𝗲𝗰𝗼𝗻𝗲, and 𝗡𝗲𝗼𝟰𝗷 offer scalable infrastructure to handle large-scale, heterogeneous knowledge. 𝗕𝗼𝗻𝘂𝘀 𝗟𝗮𝘆𝗲𝗿: 𝗜𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗶𝗼𝗻 & 𝗥��𝗮𝘀𝗼𝗻𝗶𝗻𝗴 → Integrate third-party tools via APIs → Use 𝗠𝗖𝗣 (𝗠𝘂𝗹𝘁𝗶-𝗖𝗼𝗺𝗽𝗼𝗻𝗲𝗻𝘁 𝗣𝗿𝗼𝘁𝗼𝗰𝗼𝗹) 𝘀𝗲𝗿𝘃𝗲𝗿𝘀 for orchestration → Implement custom 𝗿𝗲𝗮𝘀𝗼𝗻𝗶𝗻𝗴 𝗳𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸𝘀 to enable task decomposition, planning, and decision-making Whether you're building a personal AI assistant, autonomous agent, or enterprise-grade GenAI solution—𝘀𝗰𝗮𝗹𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗱𝗲𝗽𝗲𝗻𝗱𝘀 𝗼𝗻 𝘁𝗵𝗼𝘂𝗴𝗵𝘁𝗳𝘂𝗹 𝗱𝗲𝘀𝗶𝗴𝗻 𝗰𝗵𝗼𝗶𝗰𝗲𝘀, 𝗻𝗼𝘁 𝗷𝘂𝘀𝘁 𝗯𝗶𝗴𝗴𝗲𝗿 𝗺𝗼𝗱𝗲𝗹𝘀. Are you using these components in your architecture today?

Most corporate incubators fail. Here's why, and how to beat the odds. After watching countless innovation labs launch with fanfare only to quietly disappear, I've identified patterns that separate winners from the "Innovation Man in tights" disasters. 10 Principles for Building Incubators That Actually Work: 1. Think Like a VC, Not a Planner -- Expect 6 out of 10 projects to fail completely. If everything succeeds, you're not taking enough risk. 2. Build a Portfolio, Not Pet Projects -- Spread bets across multiple ventures. One executive's favorite idea ≠ a diversified innovation strategy. 3. Separate the Budget -- Don't make business units choose between today and tomorrow. The corporation must make that choice. 4. Focus on One Risk at a Time -- Stop trying to solve everything simultaneously. Address the biggest, cheapest-to-test risk first. 5. De-emphasize the Revenue Projections -- As one VC told me: "We look at the revenue picture, then throw it out the window. It's a dream at best." 6. Create Cross-Functional Governance Early -- Waiting until commercialization to involve other departments is a recipe for political disaster. 7. Partner with "Ball Catchers" -- Involve the people who'll eventually scale your venture from day one, even if it's just a small allocation of their time. 8. Measure "Return on Knowledge" -- When ventures are 2+ years from launch, track learnings, not just financials. 9. Awards, Not Rewards -- Entrepreneur-style financial incentives create internal warfare. Recognition works better. 10. Move Fast on Setup -- Spend 4 months max designing your program. Then launch and let real projects set the precedent. The hardest truth is that culture follows systems, not posters or people in costumes. Change your processes first, and the innovative mindset will follow.

Warning: playing can deliver serious results Sports are one area where we adults allow ourselves to play But we tend to underestimate play in other spheres. We forget play is how we make sense of the world At work, home or outside. Children are generally much more imaginative and open-minded Because they don’t believe they have all the answers And they are willing to experiment without ego. If you are trying to solve a complex problem Try these techniques: 1️⃣ Play around with it (set a time limit) 2️⃣ Subvert the script (do the opposite of what you normally do) 3️⃣ Improvise (accept and build on others' ideas instead of opposing) 4️⃣ Test and use feedback 5️⃣ Look at it from various angles with different hats on 6️⃣ Reverse engineer a solution from a desired outcome 7️⃣ Use drawing, role-play, ideation cards, dice, blocks or other playful means 8️⃣ Play the fool or devil’s advocate 9️⃣ Make no assumptions, question everything 🔟 Collaborate with people you would not usually ask You may feel uncomfortable or out of control But you could create a workable solution You otherwise would never have considered Play can help you reach the answer Plus, it builds relationships Encourages a growth mindset And it's fun. I use it to help create change. What's your experience? #play #creativity #problemsolvingskills

How often do we rush to solutions, only to realize later that we misunderstood the problem?  In project management, this is a common trap. The pressure to deliver quickly often overshadows the need to deeply understand the problem. But here's the truth: asking the right questions first comes the best solutions.  Here's a simple formula to guide you:   Understand + Analyze + Create = Great Solutions 1. Understand: Take the time to define the problem clearly. This is the foundation of effective problem-solving. Without it, you risk solving the wrong issue. 2. Analyze: Use techniques like the "5 Whys" or a Problem Tree to uncover root causes. These methods help you see beyond the surface and identify what's going on. 3. Create: Combine creativity with structure. Approaches like Design Thinking allow you to explore innovative solutions while staying focused on the problem at hand. The projects that succeed aren't the ones that move the fastest—they're the ones that solve the right problems. So, let's rethink how we approach challenges. Let's prioritize understanding over urgency. What's your go-to method for defining problems in your projects? Do you have a favourite framework or technique? Let's discuss. → Found this helpful? Repost ♺ to share, and follow Jesus Romero for more insights.

Yesterday we ran an internal AI hackathon for 100+ employees + executives across 10 departments at a CPG company. In 4 Mental Gym AI cohort-based sessions + 1:1 private sessions with workflow owners, we took teams from “I’ve never touched this” to ~70% adoption inside the company—because they didn’t just learn prompts. They built agents by function inside our LLM Lab and proved (to themselves) that AI can remove busywork without removing ownership. In just four sessions, we mapped their workflows end-to-end—looking at every step—to identify what could be automated and where the low-hanging fruit was. That groundwork is what made the hackathon so powerful: teams weren’t guessing. They were building directly on real workflows and real bottlenecks. What teams built (examples) 🦾 Supply Chain / Ops: an agent that turns messy notes + spreadsheets into a clean action plan (owners, deadlines, risks), and drafts vendor follow-ups. 🦾 Finance: an agent that flags anomalies in recurring reports, explains variance drivers, and generates CFO-ready narratives. 🦾 Marketing / Brand: an agent that generates campaign briefs + creative variations aligned to brand guardrails, then summarizes performance and next bets. 🦾 HR: an agent that standardizes interview feedback, drafts role scorecards, and keeps hiring comms moving without ping-pong. 🦾 Commercial / Sales: an agent that summarizes customer calls, extracts objections, and drafts follow-up sequences. The KPI lens we used (so it’s not “AI theater”) 👋🏽 Hours saved / week per role (admin + reporting time) 👋🏽 Cycle time reduction (request → output, days → hours) 👋🏽 Quality uplift (fewer errors, fewer back-and-forth revisions, clearer decisions) 👋🏽 Throughput (more analyses / proposals / briefs shipped per week) 👋🏽 Adoption (weekly active users, repeat usage by workflow, # of agents shipped per team) 👋🏽 Employee buy-in (ownership + confidence: “I can do this myself”) The most exciting part: this model gives power back to employees and executives. They don’t need engineering. They don’t need a consultant for every small workflow. They can prototype, test, iterate -- and own the impact. I’m honestly blown away by the creativity in the room. The innovation wasn’t ideas. It was working agents tied to real pain points, real minutes, real outcomes. ---- If you’re interested in running a Mental Gym AI LLM Lab by Function for your employees in 2026, DM me.

In the rapidly evolving enterprise landscape, ERP is no longer just about automation — it’s about augmentation. As AI reshapes the fabric of operational intelligence, C-Level leaders must shift from legacy systems toward architectures that think, adapt, and learn. This article explores how modern ERP systems are transforming into intelligent engines that amplify human capability, not replace it. The transition from workflows to intelligence marks a fundamental shift. No longer do systems just route tasks — they contextualize them, learn from outcomes, and recommend the best next steps. For executive leaders, this means real-time decision frameworks that are data-rich, insight-driven, and predictive by design. We spotlight how Machine Learning becomes the nervous system of the enterprise, quietly tuning forecasts, optimizing supply chains, and exposing hidden risks. Unlike traditional reporting, these new systems can anticipate and act, creating a proactive rather than reactive operational culture. Yet, we draw a clear boundary: AI doesn’t replace human judgment — it augments it. Modern ERP must serve as executive copilot, not autocrat. Systems must be designed to amplify strategy, not override it, providing clarity while leaving ultimate control in the hands of leadership. The most forward-thinking organizations understand: The future of ERP isn’t humanless — it’s Human-Plus. A future where technology empowers human intent, accelerates time-to-insight, and unlocks a new standard of organizational agility. #ERP #AIIntegration #Leadership #DecisionMaking #EnterpriseArchitecture #DigitalTransformation #HumanPlus #CLevelThinking

I was reading Jared Hall's latest article and looking forward to his Ignite session with Sameer Verma and Niels Nybo Jensen Here is My Take on “Agentic ERP: What the New Dynamics 365 MCP Servers Unlock” Jarad Hall’s breakdown of the Dynamic MCP Server and the upcoming Analytics MCP Server is one of the clearest explanations of where ERP is headed. But the real story isn’t “AI inside ERP.” It’s ERP shifting from a passive system to an active participant in the business. Here are the angles that matter most right now. 1. The Dynamic MCP Server ends the era of ERP bottlenecks The original MCP server was a fixed toolbox. Useful, but limited. The new dynamic version changes everything. Anything a human can do in the ERP UI, an agent can now do too. Operational work isn’t assisted anymore. It’s executable. Today. This elevates ERP teams from process operators to process designers—a mindset many organizations still lack. 2. Analytics MCP turns ERP agents from doers into thinkers When Analytics MCP arrives, agents gain access to KPIs, hierarchies, measures, variance, and forecasting. This shifts agents from task execution to decision support. My skeptical take: most enterprises will underestimate the governance and model quality required. Agents reflect the semantic layer they reason over. Messy BPA models equal messy agents. 3. Unified governance is the quiet revolution Everyone will hype the autonomy. The real innovation is that MCP now follows ERP security, RBAC, and audit controls. It’s deployable at scale not just because it’s powerful, but because it’s governable. 4. Partners and ISVs just got a new industry to build Most partners will think in solutions. The real advantage is agentic workflows. If you’re an ISV, your product is now agent-ready. If you’re a partner, your IP becomes a library of reusable agentic patterns. This opens the door to a new category I’m calling “Agentic ERP Packs.” 6. The winning architecture is now obvious You’ll need: • Copilot Studio or Azure AI agents for orchestration • Dynamic MCP for operational tasks • Analytics MCP for reasoning • External connectors for anything outside ERP’s domain This stack is clean, scalable, and practical—no more duct-taped automation. My conclusion: Agentic ERP isn’t a feature. It’s a new operating model. Dynamic MCP gives agents hands. Analytics MCP gives them a brain. Governance gives them guardrails. Processes aren’t coded- they’re delegated. This is the biggest shift in business systems since the move to the cloud. Treat it as an operating model change, not a workflow upgrade. Please tune into the Ignite sessions to learn from our product leaders: https://lnkd.in/gCDxBb5M

90% of transformations fail. I know why. (And how to be in the 10%.) I spent years helping CEOs create billion-dollar transformations as head coach for Satya Nadella's innovation team. Here are the principles, patterns, and tools that actually work. The biggest lesson comes down to this: 𝗜𝗻𝗻𝗼𝘃𝗮𝘁𝗶𝗼𝗻 𝗶𝘀𝗻'𝘁 𝗮𝗯𝗼𝘂𝘁 𝗹𝘂𝗰𝗸. 𝗜𝘁'𝘀 𝗮𝗯𝗼𝘂𝘁 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗮𝘁 𝘀𝗽𝗲𝗲𝗱. Every CEO I worked with faced the same challenge: how to transform while still delivering today's results. The winners had three things in common: 1. They started with business outcomes, not technology 2. They built experiments, not just strategies 3. They visualized value before investing millions Inside this playbook: • The "Business Before Technology" framework that kept us from chasing shiny objects • The C.E.O. Pattern (Customer, Employee, Operations) for focusing innovation efforts • Journey mapping techniques we used with Fortune 500s • The micro-story structure that gets executives to actually listen • How to build a culture where experiments drive billions in value This isn't theory. These are the actual tools I used at Microsoft and beyond to help leaders transform. No fluff. Just what works. What's the #1 friction point slowing your organization's transformation? Follow me, J.D. Meier, for frameworks that created billions in business value.

Creative problem solving is not only about generating ideas. It is also about accurately judging them. A recent study by Marek Urban and Kamila Urban shows that individuals who are more metacognitively accurate — better at evaluating the quality of their own ideas — also tend to perform more creatively. Even more compelling: creative metacognitive accuracy appears to transfer across tasks. Accuracy in evaluating ideas in one type of creativity task predicted performance in another. Researchers asked a critical question: Are there individuals who are highly creative in one task but metacognitively inaccurate in another? Their answer was clear: no. Creative metacognitive accuracy seems to be necessary — though not sufficient — for strong creative performance. This has important practical implications. If we want to foster creativity, we cannot focus only on idea generation. We must also help people become better judges of their own thinking. Developing creative potential means developing self-awareness about our creative performance. #Metacognition  #CreativeProcess  #ProblemSolving  #CognitiveScience

🔶 Is innovation truly driving impact in your organization, or just creating pilots that never scale? Over the years, I've observed a recurring issue: organizations execute impressive proofs-of-concept (POCs) and then they stall. The excitement fades. The pilot never scales. The value never materializes. The good news? There are ways to break free. Here are four strategies that I’ve seen work consistently: 1️⃣ Tie the POC to a business KPI from day one If an idea doesn’t connect to measurable business value, it’s unlikely to get sponsorship. 2️⃣ Bring cross-functional stakeholders in early Success requires the buy-in of more functions in the organization than IT. 3️⃣ Set a decision timeline Scale, iterate, or stop, but don’t let pilots drag on indefinitely. Clarity beats limbo. 4️⃣ Celebrate and communicate quick wins Momentum builds when leadership sees early impact. Recognition accelerates adoption. 💡 Innovation is not about experimenting endlessly, it’s about moving ideas into production where they create real business outcomes. 👉 My takeaway: The difference between a pilot and a transformation is the ability to scale with purpose. I’m curious: What’s one tactic you’ve used (or seen) that helped a pilot succeed and scale?